With a background of close to 25 years in DevOps and more than 15 years of in-depth experience in DevOps security, mostly in the financial and the public sector, John is extremely well positioned to frame IT security in strongly regulated environments. Recently, this exposure increasingly included AI-related security and governance topics, providing him we the necessary insights into the specifics of a fluid and non-deterministic environment.

His recent approach to AI was to deliberately move to extensive experimental AI work, which has given him an exhaustive understanding of the specifics of an AI environment and its particular challenges.

That work trains a particular reflex: assume the system is wrong until it proves otherwise, and don't confuse a demo with something that survives production.

He started turning that reflex on AI with small software projects and good results — commercially ready and source-available. His code is gated on SAST and DAST security reviews before publication, with documented triage decisions.

John's emerging AI work covers security and operational integrity within AI systems, including key and secret handling, supply chain, deployment hardening, and what evidence survives for future audits.

He advocates an "arbitrage master" role for entry-level junior engineers using multiple AI models to add value and gain work experience, rather than being "replaced" by AI in the first place.

John often uses "AI through humor" to explain topics, keeping the human element in view.

John obtained a BsC in Computer Science & Engineering at the University of Michigan and has extensive certified knowledge in the areas relevant to secure DevOps environments.